EFFECT OF EVALUATOR AND RESIDENT GENDER ON THE AMERICAN-BOARD OF INTERNAL-MEDICINE EVALUATION SCORES

OBJECTIVE: To examine the effects of resident and attending physician gender on the evaluation of residents in an internal medicine training program. DESIGN: Cross-sectional study. SETTING: Large urban academic internal medicine residency program. PARTICIPANTS: During their first 2 years of training, 132 residents (85 men, 47 women) received a tota l of 974 evaluations from 255 attending physicians (203 men, 52 women) from 1989 to 1995. MEASUREMENTS: The primary measurements were the nu merical portions of the American Board of Internal Medicine evaluation form. Separate analyses were performed for each of the nine evaluatio n dimensions graded on a scale of 1 to 9. The primary outcome was the difference in the average scores received by each resident from male v ersus female attending physicians. RESULTS: Compared with female train ees, male residents received significantly higher scores from male att ending physicians than from female attending physicians in six of the nine dimensions: clinical judgment, history, procedures, relationships , medical care, and overall. Similar trends, not reaching conventional levels of statistical significance, were observed in the other three categories: medical knowledge, physical exam, and attitude. These diff erences ranged from 0.24 to 0.60 points, and were primarily due to hig her grading of male residents by male attending physicians than by fem ale attending physicians. CONCLUSIONS: In one academic training progra m, we found a significant interaction in the grading process between t he gender of internal medicine residents and the gender of their atten ding evaluators. This study raises the possibility that subtle aspects of gender bias may exist in medical training programs.